The recent, rapid development and popularization of Large Language Models (LLM) have transformed the panorama of Natural Language Processing (NLP) and, more generally, of Artificial Intelligence (AI), permeating into society and transforming the way many tasks are performed, being now either supported or automated with the help of LLM-based tools. Along with the challenges of hallucinations, lack of reasoning capabilities, inability to perform numerical calculations, natural aging of the training data, and improper traceability and citation of information sources, another intrinsic challenge of LLMs, tightly related to their architecture and training, concerns their limited context window and maximum token output (Kaddour et al., 2023). Indeed, the context window is the cornerstone for LLM-based applications which require the previous interactions in the conversation to be preserved and considered by the LLM. This, being true for long conversations, is of particular importance in the engineering design field when an LLM is used to support engineers in the design of a system, going from high-level concept generation to lower-level system requirements or technical specifications generation. This application requires previous decisions as well as the decision-making process to be considered in later stages.

This paper addresses the challenge of accurately translating technical terms, which are crucial for clear communication in specialized fields. We introduce the Parenthetical Terminology Translation (PTT) task, designed to mitigate potential inaccuracies by displaying the original term in parentheses alongside its translation. To implement this approach, we generated a representative PTT dataset using a collaborative approach with large language models and applied knowledge distillation to fine-tune traditional Neural Machine Translation (NMT) models and small-sized Large Language Models (sLMs). Additionally, we developed a novel evaluation metric to assess both overall translation accuracy and the correct parenthetical presentation of terms. Our findings indicate that sLMs did not consistently outperform NMT models, with fine-tuning proving more effective than few-shot prompting, particularly in models with continued pre-training in the target language. These insights contribute to the advancement of more reliable terminology translation methodologies.

This paper presents the Language Aided Subset Sampling Based Motion Planner (LASMP), a system that helps mobile robots plan their movements by using natural language instructions. LASMP uses a modified version of the Rapidly Exploring Random Tree (RRT) method, which is guided by user-provided commands processed through a language model (RoBERTa). The system improves efficiency by focusing on specific areas of the robot's workspace based on these instructions, making it faster and less resource-intensive. Compared to traditional RRT methods, LASMP reduces the number of nodes needed by 55% and cuts random sample queries by 80%, while still generating safe, collision-free paths. Tested in both simulated and real-world environments, LASMP has shown better performance in handling complex indoor scenarios. The results highlight the potential of combining language processing with motion planning to make robot navigation more efficient.

Text style transfer (TST) aims to modify the style of a text without altering its original meaning. Large language models (LLMs) demonstrate superior performance across multiple tasks, including TST. However, in zero-shot setups, they tend to directly copy a significant portion of the input text to the output without effectively changing its style. To enhance the stylistic variety and fluency of the text, we present sNeuron-TST, a novel approach for steering LLMs using style-specific neurons in TST. Specifically, we identify neurons associated with the source and target styles and deactivate source-style-only neurons to give target-style words a higher probability, aiming to enhance the stylistic diversity of the generated text. However, we find that this deactivation negatively impacts the fluency of the generated text, which we address by proposing an improved contrastive decoding method that accounts for rapid token probability shifts across layers caused by deactivated source-style neurons. Empirical experiments demonstrate the effectiveness of the proposed method on six benchmarks, encompassing formality, toxicity, politics, politeness, authorship, and sentiment.

The impressive performance of proprietary LLMs like GPT4 in code generation has led to a trend to replicate these capabilities in open-source models through knowledge distillation (e.g. Code Evol-Instruct). However, these efforts often neglect the crucial aspect of response quality, relying heavily on teacher models for direct response distillation. This paradigm, especially for complex instructions, can degrade the quality of synthesized data, compromising the knowledge distillation process. To this end, our study introduces the Adaptive Modular Response Evolution (AMR-Evol) framework, which employs a two-stage process to refine response distillation. The first stage, modular decomposition, breaks down the direct response into more manageable sub-modules. The second stage, adaptive response evolution, automatically evolves the response with the related function modules. Our experiments with three popular code benchmarks (HumanEval, MBPP, and EvalPlus) attest to the superiority of the AMR-Evol framework over baseline response distillation methods. By comparing with the open-source Code LLMs trained on a similar scale of data, we observed performance enhancements: more than +3.0 points on HumanEval-Plus and +1.0 points on MBPP-Plus, which underscores the effectiveness of our framework. Our codes are available at https://github.com/ChiYeungLaw/AMR-Evol.

Large model inference is shifting from cloud to edge due to concerns about the privacy of user interaction data. However, edge devices often struggle with limited computing power, memory, and bandwidth, requiring collaboration across multiple devices to run and speed up LLM inference. Pipeline parallelism, the mainstream solution, is inefficient for single-user scenarios, while tensor parallelism struggles with frequent communications. In this paper, we argue that tensor parallelism can be more effective than pipeline on low-resource devices, and present a compute-and memory-efficient tensor parallel inference system, named TPI-LLM, to serve 70B-scale models. TPI-LLM keeps sensitive raw data local in the users' devices and introduces a sliding window memory scheduler to dynamically manage layer weights during inference, with disk I/O latency overlapped with the computation and communication. This allows larger models to run smoothly on memory-limited devices. We analyze the communication bottleneck and find that link latency, not bandwidth, emerges as the main issue, so a star-based allreduce algorithm is implemented. Through extensive experiments on both emulated and real testbeds, TPI-LLM demonstrated over 80% less time-to-first-token and token latency compared to Accelerate, and over 90% compared to Transformers and Galaxy, while cutting the peak memory footprint of Llama 2-70B by 90%, requiring only 3.1 GB of memory for 70B-scale models. Recently, Large Language Models (LLMs) have been widely deployed in the cloud for inference. User inputs are uploaded to the cloud, where high-performance GPUs are used to compute output sequences, and then sent back to user devices for display. This process poses privacy risks, as user prompts are exposed to network intermediaries and clouds.

Instructional documents are rich sources of knowledge for completing various tasks, yet their unique challenges in conversational question answering (CQA) have not been thoroughly explored. Existing benchmarks have primarily focused on basic factual question-answering from single narrative documents, making them inadequate for assessing a model`s ability to comprehend complex real-world instructional documents and provide accurate step-by-step guidance in daily life. To bridge this gap, we present InsCoQA, a novel benchmark tailored for evaluating large language models (LLMs) in the context of CQA with instructional documents. Sourced from extensive, encyclopedia-style instructional content, InsCoQA assesses models on their ability to retrieve, interpret, and accurately summarize procedural guidance from multiple documents, reflecting the intricate and multi-faceted nature of real-world instructional tasks. Additionally, to comprehensively assess state-of-the-art LLMs on the InsCoQA benchmark, we propose InsEval, an LLM-assisted evaluator that measures the integrity and accuracy of generated responses and procedural instructions.

Learning a model of a stochastic setting often involves learning both general structure rules and specific properties of the instance. This paper investigates the interplay between learning the general and the specific in various learning methods, with emphasis on sample efficiency. We design a framework called {\sc LeverWorlds}, which allows the generation of simple physics-inspired worlds that follow a similar generative process with different distributions, and their instances can be expressed in natural language. These worlds allow for controlled experiments to assess the sample complexity of different learning methods. We experiment with classic learning algorithms as well as Transformer language models, both with fine-tuning and In-Context Learning (ICL). Our general finding is that (1) Transformers generally succeed in the task; but (2) they are considerably less sample efficient than classic methods that make stronger assumptions about the structure, such as Maximum Likelihood Estimation and Logistic Regression. This finding is in tension with the recent tendency to use Transformers as general-purpose estimators. We propose an approach that leverages the ICL capabilities of contemporary language models to apply simple algorithms for this type of data. Our experiments show that models currently struggle with the task but show promising potential.

Cross-lingual summarization aims to bridge language barriers by summarizing documents in different languages. However, ensuring semantic coherence across languages is an overlooked challenge and can be critical in several contexts. To fill this gap, we introduce multi-target cross-lingual summarization as the task of summarizing a document into multiple target languages while ensuring that the produced summaries are semantically similar. We propose a principled re-ranking approach to this problem and a multi-criteria evaluation protocol to assess semantic coherence across target languages, marking a first step that will hopefully stimulate further research on this problem.

Large Language Models (LLMs) require frequent updates to correct errors and keep pace with continuously evolving knowledge in a timely and effective manner. Recent research in it model editing has highlighted the challenges in balancing generalization and locality, especially in the context of lifelong model editing. We discover that inserting knowledge directly into the model often causes conflicts and potentially disrupts other unrelated pre-trained knowledge. To address this problem, we introduce UniAdapt, a universal adapter for knowledge calibration. Inspired by the Mixture of Experts architecture and Retrieval-Augmented Generation, UniAdapt is designed with a vector-assisted router that is responsible for routing inputs to appropriate experts. The router maintains a vector store, including multiple shards, to construct routing vectors based on semantic similarity search results. UniAdapt is fully model-agnostic and designed for seamless plug-and-play integration. Experimental results show that UniAdapt outperforms existing lifelong model editors and achieves exceptional results in most metrics.